Announcement

Collapse
No announcement yet.

DNA signature for half-sisters

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • DNA signature for half-sisters

    What effect would a half-sister vs full-sister relationship have on Family Finder results at the level of first cousins?

    In my case two women who believe their mothers were full-sisters used Family Finder with match results of total cM: 885.06 and longest block: 96.67.

    Do these numbers rule out a half-sister relationship for their mothers? i.e. same father different mothers for the sisters? Or is that still a possibility? Is there a table of number-ranges for various relationships somewhere handy?

    Rob

  • #2
    My understanding is that full siblings should half-match at about 75% & half siblings should half match at about 50%. Full first cousins should have match at about 25%. Half first cousins should half match at 12.5%.

    This is how I understand the numbers to work. Correct me if I'm wrong.

    Timothy Peterman

    Comment


    • #3
      Originally posted by robdrew7 View Post
      What effect would a half-sister vs full-sister relationship have on Family Finder results at the level of first cousins?

      In my case two women who believe their mothers were full-sisters used Family Finder with match results of total cM: 885.06 and longest block: 96.67.

      Do these numbers rule out a half-sister relationship for their mothers? i.e. same father different mothers for the sisters? Or is that still a possibility? Is there a table of number-ranges for various relationships somewhere handy?

      Rob
      Timothy quoted some average numbers for half-identical sharing, but in real life, you can end up in a fuzzy gray area. I know a case where two known first cousins are classified differently at FTDNA. In one case, the suggested relationship is 1st cousin and in the other case it is 2nd cousin with a range of 1st to 2nd (consistent with half-first cousins). The "2nd cousin" just happened to inherit less than the average amount, purely by chance. I have a similar case at 23andMe -- I match a 1st cousin pretty much as "expected", but my sister inherited fewer segments and is actually closer to the expected numbers for a 1st cousin once removed (or half-first cousin). The range for my sister is given as 1st to 2nd cousin.

      FWIW, your numbers are closer to the one predicted as a 1st cousin at FTDNA. What does FTDNA say?

      Comment


      • #4
        My use of the word "about" ws suggestive of this gray area. Siblings could share anywhere from 0% to 100% of their DNA, although unless identical, the outliers of 0 & 100 would be rather rare. If siblings average at 50% shared DNA, the same siblings would, on average, half match at 75%.

        Timothy Peterman

        Comment


        • #5
          Originally posted by Ann Turner View Post
          FWIW, your numbers are closer to the one predicted as a 1st cousin at FTDNA. What does FTDNA say?
          Hi Ann, FTDNA predicted Aunt/Uncle to Niece/Nephew or 1st cousins. When you and Timothy quote %'s, I'm wondering if 100% is about 3400 cM, making this about 25% as predicted for 1st cousins?

          Rob

          Comment


          • #6
            When I compare myself to my dad in chromosome browser, every bit of the compared parts of the 22 relevant chromosomes are lit up. I describe this as half matching at 100%. I think the shared cm count is something like 3379 or near there.

            Timothy Peterman

            Comment


            • #7
              This is a higher than average result for first cousins. I think it would be very unusual to find in half first cousins, unless there were other relationships (for example, the two fathers were brothers).

              I am looking at a list of about 50 known 1st cousins that was published on this forum. The total DNA ranges from about 550 to 1150cM. In my family I have three known first cousin relationships and the total dna is 748, 823 and 1128cM for those. Except for one longest segment listed at 128cM (which I consider must be an unusual situation), the longest segments in the list of 50 1st cousins go from about 40 to 81 cM. So your longest segment would be at the top of that list.

              Looks like full first cousins.

              Comment


              • #8
                Based on your numbers, I'm going with full-cousins. Thank you!

                Rob

                Comment

                Working...
                X